1. Field of the Invention
The present invention relates generally to analytics for software development. Further, the present invention relates to gathering data and comparing it with best practices to establish status of a project and corrective actions.
2. Description of the Prior Art
The software development lifecycle is complex. It comprises multiple phases that are carried out sequentially or in parallel. These phases require teams of individuals that are often located in different physical locations to co-ordinate with each other in an efficient manner to complete all the activities in a given time. Estimating the time required for the software development phases is a harrowing task that all project managers have difficulty coping with.
It is even more difficult is to estimate the amount of time spent on a particular task in a particular phase of a software development project. This is because team members often work on multiple software development projects simultaneously. It is important to have an accurate estimate of the time spent on a project by all team members in order to accurately estimate the time remaining and the completion date of the project. It also helps in providing managers with knowledge about the current status of the project, and if the project is proceeding as planned.
Project managers estimate time spent on a project by requesting the team members to fill out weekly or daily timesheets. The team members fill these time sheets from memory, and hence these do not provide an accurate estimate of the time spent on the various activities of a project. This in turn leads to an inaccurate picture about the current status of a project. Without a clear picture about the current status of a project, it becomes even more difficult to predict the completion date for a project and adhere to it.
It is also difficult to estimate the performance of teams with such inaccurate data. One measure of performance is comparison of the team's performance with industry wide best practices. Since the best practices are computed by collecting performance data, which is inaccurate to begin with, from multiple sources, the comparison will also not yield an accurate picture of the state of the current processes.
There are numerous solutions in the prior art addressed towards this problem. One such example is U.S. Pat. No. 6,519,763 titled “Time Management and Task Completion and Management Software” assigned to Compuware Corporation. This patent describes an apparatus for ascertaining project completion and managing a project with high efficiency and accuracy. It comprises of data collectors that automatically gather data that is generated by various tools, such as scheduling, defect tracking, and other software management and quality tools. The data is analyzed to generate statistical measures relating to the status of the project. The data collectors collect data from project management software, defect reports and testing reports and other sources that provide information about the project status.
Another such solution is described in US Patent Application Pub. No. 20050289503 titled “System for Identifying Project Status and Velocity through Predictive Measures” by Clifford, Gregory. This patent application describes a method for providing visibility into the real time progress and status of software development projects by collecting measures from software development tools about the progress of the project, examining data sources created during the progress of the project and evaluating the collected data by using expert reasoning system based on causal modeling to arrive at project velocity views. The system collects data from sources such as configuration management systems, defect management systems, project management system and source code.
Although the prior art tried to address the problem of estimating project completion and project status accurately, it fails to address the need for a system that compares project status information with best practices in the industry. Further, the prior art does not describe a way in which project completion information from various companies' offices can be stored to generate best practices without compromising the security of the data. Additionally, the prior art does not describe a way in which data is collected from various software development tools and using that data to estimate project completion and real time status. Collecting data from software development tools provides insights into the software development process and one can compare the time a person spends on various tools for completing a given task. This provides insight into a deeper level that the prior art, in general, does not provide.
From the above discussion, it is clear that there is a need for a system that will collect data from tools used for software development, but will also collect data across organizations and store the data in such a way that privacy of the organizations is not compromised. Further, the system should be able to generate best practices from the collected data and to provide comparisons about current trends of a software project with the industry best practices.